Large-scale machine-learning-assisted exploration of the whole materials space


JSON Export

{
  "revision": 4, 
  "metadata": {
    "publication_date": "Oct 04, 2022, 10:24:01", 
    "_oai": {
      "id": "oai:materialscloud.org:1485"
    }, 
    "license": "Creative Commons Attribution 4.0 International", 
    "description": "Crystal-graph attention networks have emerged recently as remarkable tools for the prediction of thermodynamic stability and materials properties from unrelaxed crystal structures. Previous networks trained on two million materials exhibited, however, strong biases originating from underrepresented chemical elements and structural prototypes in the available data. We tackled this issue computing additional data to provide better balance across both chemical and crystal-symmetry space. Crystal-graph networks trained with this new data show unprecedented generalization accuracy, and allow for reliable, accelerated exploration of the whole space of inorganic compounds. We applied this universal network to performed machine-learning assisted high-throughput materials searches including 2500 binary and ternary prototypes and spanning about 1 billion compounds. After validation using density-functional theory, we uncover in total 19512 additional materials on the convex hull of thermodynamic stability and around 150000 compounds with a distance of less than 50 meV/atom from the hull. Here we include the DCGAT-1, DCGAT-2, and DCGAT-3 datasets used in this work.", 
    "contributors": [
      {
        "familyname": "Schmidt", 
        "affiliations": [
          "Institut f\u00fcr Physik, Martin-Luther-Universit\u00e4t Halle-Wittenberg, 06120 Halle (Saale), Germany."
        ], 
        "givennames": "Jonathan"
      }, 
      {
        "familyname": "Hoffmann", 
        "affiliations": [
          "Institut f\u00fcr Physik, Martin-Luther-Universit\u00e4t Halle-Wittenberg, 06120 Halle (Saale), Germany."
        ], 
        "givennames": "Noah"
      }, 
      {
        "familyname": "Wang", 
        "affiliations": [
          "Institut f\u00fcr Physik, Martin-Luther-Universit\u00e4t Halle-Wittenberg, 06120 Halle (Saale), Germany."
        ], 
        "givennames": "Hai-Chen"
      }, 
      {
        "familyname": "Borlido", 
        "affiliations": [
          "CFisUC, Department of Physics, University of Coimbra, Rua Larga, 3004-516 Coimbra, Portugal"
        ], 
        "givennames": "Pedro"
      }, 
      {
        "familyname": "M.A. Carri\u00e7o", 
        "affiliations": [
          "CFisUC, Department of Physics, University of Coimbra, Rua Larga, 3004-516 Coimbra, Portugal"
        ], 
        "givennames": "Pedro J."
      }, 
      {
        "familyname": "F. T. Cerqueira", 
        "affiliations": [
          "CFisUC, Department of Physics, University of Coimbra, Rua Larga, 3004-516 Coimbra, Portugal"
        ], 
        "givennames": "Tiago"
      }, 
      {
        "familyname": "Botti", 
        "affiliations": [
          "Institut f\u00fcr Festk\u00f6rpertheorie und -optik and European Theoretical Spectroscopy Facility, Friedrich-Schiller-Universit\u00e4t Jena, D-07743 Jena, Germany"
        ], 
        "email": "silvana.botti@uni-jena.de", 
        "givennames": "Silvana"
      }, 
      {
        "familyname": "L. Marques", 
        "affiliations": [
          "Institut f\u00fcr Physik, Martin-Luther-Universit\u00e4t Halle-Wittenberg, 06120 Halle (Saale), Germany."
        ], 
        "email": "miguel.marques@physik.uni-halle.de", 
        "givennames": "Miguel A."
      }
    ], 
    "edited_by": 578, 
    "title": "Large-scale machine-learning-assisted exploration of the whole materials space", 
    "conceptrecid": "1484", 
    "license_addendum": null, 
    "doi": "10.24435/materialscloud:m7-50", 
    "mcid": "2022.126", 
    "_files": [
      {
        "size": 400, 
        "key": "test_json.py", 
        "checksum": "md5:544902c43b476ed5c7e0c8d3ce338365", 
        "description": "Example program to load the data"
      }, 
      {
        "size": 4542, 
        "key": "README.txt", 
        "checksum": "md5:0691ad62b6b20382a64cdaf78b702e14", 
        "description": "Detailed description"
      }, 
      {
        "size": 31893301, 
        "key": "dcgat_1_000.json.bz2", 
        "checksum": "md5:4800e5353f3a663cae973bd3c0397c76", 
        "description": "DGCAT-1-000"
      }, 
      {
        "size": 31417703, 
        "key": "dcgat_1_001.json.bz2", 
        "checksum": "md5:342110c4d7194f1a7b4bd5589cbb790a", 
        "description": "DGCAT-1-001"
      }, 
      {
        "size": 32526640, 
        "key": "dcgat_1_002.json.bz2", 
        "checksum": "md5:f6161fdbcaed5d949bc41310002882f9", 
        "description": "DGCAT-1-002"
      }, 
      {
        "size": 34987444, 
        "key": "dcgat_1_003.json.bz2", 
        "checksum": "md5:5281d8c75f2b780eae25a8df51ba80a6", 
        "description": "DGCAT-1-003"
      }, 
      {
        "size": 31169129, 
        "key": "dcgat_1_004.json.bz2", 
        "checksum": "md5:5b6cb04653a7b55b868f34fb237717e9", 
        "description": "DGCAT-1-004"
      }, 
      {
        "size": 33012760, 
        "key": "dcgat_1_005.json.bz2", 
        "checksum": "md5:db570fc7f685059f71288a0c99e2018a", 
        "description": "DGCAT-1-005"
      }, 
      {
        "size": 36584225, 
        "key": "dcgat_1_006.json.bz2", 
        "checksum": "md5:3360ad70980a88ca9ffd05b4d9a3d6c4", 
        "description": "DGCAT-1-006"
      }, 
      {
        "size": 33078830, 
        "key": "dcgat_1_007.json.bz2", 
        "checksum": "md5:1d338c01c7a8cfc53f40259869f3d382", 
        "description": "DGCAT-1-007"
      }, 
      {
        "size": 33326402, 
        "key": "dcgat_1_008.json.bz2", 
        "checksum": "md5:3f4a55ca617741e5a4a32d861037c2b4", 
        "description": "DGCAT-1-008"
      }, 
      {
        "size": 34102019, 
        "key": "dcgat_1_009.json.bz2", 
        "checksum": "md5:59961ff20b1b2199e8c63badc12f95b3", 
        "description": "DGCAT-1-009"
      }, 
      {
        "size": 30643306, 
        "key": "dcgat_1_010.json.bz2", 
        "checksum": "md5:536ea5b36fca25278464ffa30d90f225", 
        "description": "DGCAT-1-010"
      }, 
      {
        "size": 32841524, 
        "key": "dcgat_1_011.json.bz2", 
        "checksum": "md5:19e1755fa8fd4bef9e241dfee2ddb6dd", 
        "description": "DGCAT-1-011"
      }, 
      {
        "size": 32475973, 
        "key": "dcgat_1_012.json.bz2", 
        "checksum": "md5:3a05fc9bbb90b20ea1a144ac158c7cf4", 
        "description": "DGCAT-1-012"
      }, 
      {
        "size": 32826119, 
        "key": "dcgat_1_013.json.bz2", 
        "checksum": "md5:0357c6e0a16ab1fd56075aa013d5afe5", 
        "description": "DGCAT-1-013"
      }, 
      {
        "size": 14913487, 
        "key": "dcgat_1_014.json.bz2", 
        "checksum": "md5:1eb85737cab59cfff23539e5dc1f426b", 
        "description": "DGCAT-1-014"
      }, 
      {
        "size": 33694641, 
        "key": "dcgat_2_000.json.bz2", 
        "checksum": "md5:3473538c5ae6b43b82577eec5cd6a522", 
        "description": "DGCAT-2-000"
      }, 
      {
        "size": 32684402, 
        "key": "dcgat_2_001.json.bz2", 
        "checksum": "md5:33da8855afabeba547c11401bbb1a07f", 
        "description": "DGCAT-2-001"
      }, 
      {
        "size": 31430691, 
        "key": "dcgat_2_002.json.bz2", 
        "checksum": "md5:f3c6e3894ebe3464413f60472d3e7dfb", 
        "description": "DGCAT-2-002"
      }, 
      {
        "size": 30694952, 
        "key": "dcgat_2_003.json.bz2", 
        "checksum": "md5:49ccc949574a5900c4e611b993b20b0d", 
        "description": "DGCAT-2-003"
      }, 
      {
        "size": 65720185, 
        "key": "dcgat_3_000.json.bz2", 
        "checksum": "md5:b8203c162da342f69733e96dfd5f19b6", 
        "description": "DGCAT-3-000"
      }, 
      {
        "size": 66599788, 
        "key": "dcgat_3_001.json.bz2", 
        "checksum": "md5:b3b564f3868962d3e70158c9e8f70a2e", 
        "description": "DGCAT-3-001"
      }, 
      {
        "size": 66221735, 
        "key": "dcgat_3_002.json.bz2", 
        "checksum": "md5:4e4f25d1a0fad143b5c3b40873d476b7", 
        "description": "DGCAT-3-002"
      }, 
      {
        "size": 65898941, 
        "key": "dcgat_3_003.json.bz2", 
        "checksum": "md5:618559c17c20717f2dae514ca953be8f", 
        "description": "DGCAT-3-003"
      }, 
      {
        "size": 66157553, 
        "key": "dcgat_3_004.json.bz2", 
        "checksum": "md5:bff5cb86fe7dfbff5941ccf70eb0be2b", 
        "description": "DGCAT-3-004"
      }, 
      {
        "size": 10091806, 
        "key": "dcgat_3_005.json.bz2", 
        "checksum": "md5:6060a9f68aa97b1b31a9b0054a349795", 
        "description": "DGCAT-3-005"
      }
    ], 
    "id": "1485", 
    "keywords": [
      "density-functional theory", 
      "high-throughput", 
      "crystal-graph attention networks"
    ], 
    "is_last": true, 
    "status": "published", 
    "references": [
      {
        "url": "https://arxiv.org/abs/2210.00579", 
        "type": "Preprint", 
        "citation": "J. Schmidt, N. Hoffmann, H.-C. Wang, P. Borlido, P. J. M. A. Carri\u00e7o, T. F. T. Cerqueira, S. Botti, M. A. L. Marques, arXiv:2210.00579 (2022)"
      }
    ], 
    "version": 1, 
    "owner": 364
  }, 
  "id": "1485", 
  "created": "2022-09-29T08:33:30.168076+00:00", 
  "updated": "2022-10-04T08:24:01.710590+00:00"
}